a new moral hazard? military intervention, peacekeeping and the
TRANSCRIPT
Eric NeumayerA new moral hazard? Military intervention, peacekeeping and the International Criminal Court Article (Accepted version) (Refereed) Original citation: Neumayer, Eric (2009) A new moral hazard? Military intervention, peacekeeping and the International Criminal Court. Journal of peace research, 46 (5). pp. 659-670. ISSN 1460-3578 DOI: 10.1177/0022343309339246 © 2009 The author; published by Sage Publications This version available at: http://eprints.lse.ac.uk/20383/ Available in LSE Research Online: September 2010 LSE has developed LSE Research Online so that users may access research output of the School. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Users may download and/or print one copy of any article(s) in LSE Research Online to facilitate their private study or for non-commercial research. You may not engage in further distribution of the material or use it for any profit-making activities or any commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE Research Online website. This document is the author’s final manuscript accepted version of the journal article, incorporating any revisions agreed during the peer review process. Some differences between this version and the published version may remain. You are advised to consult the publisher’s version if you wish to cite from it.
1
A New Moral Hazard? Military Intervention, Peacekeeping
and the International Criminal Court
Revised Version: August 2007
Eric Neumayer
London School of Economics and Political Science, Houghton Street, London WC2A
2AE, UK and Center for the Study of Civil War, International Peace Research
Institute Oslo, Norway. Phone: +44-207-9557598. Fax: +44-207-9557412. Email:
Website: http://personal.lse.ac.uk/neumayer
I thank three referees as well as Nils Petter Gleditsch for very helpful comments.
2
Abstract
The newly established International Criminal Court (ICC) promises justice to the
victims of genocide, war crimes and crimes against humanity. Past offenders can be
punished, while future potential offenders may be deterred by the prospect of
punishment. Yet, justice is no substitute for intervention for the benefit of people at
acute risk of being victimized. The Court may create a new moral hazard problem if
the promise of ex post justice makes it easier for states to shy away from incurring the
costs of intervention. This article indirectly tests for the relevance of this potential
problem by estimating the determinants of ratification delay to the Rome Statute of
the ICC. I find that countries, which in the past have been more willing to intervene in
foreign civil wars and more willing to contribute troops to multinational peacekeeping
missions are more likely to have ratified the Statute (early on). This suggests that the
Court is a complement to, not a substitute for intervention.
1. Introduction
The creation of an International Criminal Court (ICC) has been hailed by many as a
major breakthrough in international law. Weller (2002: 693), for example, has called
it ‘the culmination of international law-making of the twentieth century’. Former
United Nations (UN) General Secretary Kofi Annan regards it as ‘a vital part of an
emerging system of international human rights protection’ (Annan 1997/98: 365).
Yet, the ICC is no substitute for humanitarian intervention and multinational
peacekeeping since at best it provides ex post justice together with the hope of
deterrence of future crimes, but no immediate relief and assistance to people at risk of
becoming the victims of grave offences. Critics have argued that the ICC might just
be a cheap way for states, which are unwilling or unable to intervene militarily abroad
3
and to contribute to peacekeeping, to demonstrate to their domestic publics seeming
action against genocide, war crimes, aggression and crimes against humanity (Smith
2002). In other words, is the ICC a form of organized hypocrisy behind which states
hide who cannot or do not want to take real action?
This article attempts to answer this question indirectly by examining ratification
of the Rome Statute of the ICC. Specifically, we test whether countries that have been
more willing in the past to intervene militarily in conflicts abroad and who have
contributed more often troops to multinational peacekeeping are more or less likely to
have ratified the Rome Statute (early on). An answer to this question not only sheds
light on the likely motivations of state parties, it also informs a judgement of the
likely future effectiveness of the Court. The ICC is only as strong as its enforcement
capacity and it is dependent on states for crucial assistance during all stages of its
proceedings, particularly for the capture and extradition of indicted individuals. If
those states that are willing to commit troops for military intervention and
peacekeeping abroad were less likely to support the ICC, then this would bode badly
for the Court’s enforcement capacity.
In brief, the results from a stratified proportional hazard model over the period
1998 to 2005 suggest that, despite the notable opposition to the Court from the United
States, Russia, India, China and others, states that have been more willing to intervene
abroad and more willing to contribute to multinational peacekeeping in the past have
been more, not less, likely to ratify the Rome Statute (early on). Taking ratification as
an indication of revealed preference for ICC support, this suggests that the Court is
unlikely to function as a sorry excuse for inactive governments.
4
2. The Rome Statute of the International Criminal Court
The idea for an ICC equivalent was briefly floated in the post-Second World War era,
but soon dumped in the wake of Cold War antagonism. It was reinvigorated by
Trinidad and Tobago in the late 1980s with the support of other Caribbean countries,
which wanted to see international drug trafficking and money laundering being
prosecuted by an international court. The International Law Commission, a UN expert
body of legal scholars, was charged with formulating a draft statute, a final version of
which was presented to the UN General Assembly in 1994 (Schabas 2004).
Encouraged by the establishment of the two ad hoc tribunals for Yugoslavia and
Rwanda, the General Assembly created a preparatory committee whose objective was
to formulate a draft treaty with much wider coverage and competence for a permanent
court to be established than initially envisioned by Trinidad and Tobago or the
International Law Commission.
At a conference in the Italian capital, the Rome Statute of the ICC was opened
for signature in July 1998 after five weeks of negotiation. International civil society,
in the form of the NGO Coalition for an International Criminal Court (CICC), exerted
an important influence on the negotiations of the Rome Statute, but only after the
decision to establish an ICC in principle had already been taken by states (Fehl 2004:
374). The Rome Statute entered into force on 1 July 2002 after having been ratified by
60 state parties. As of July 2007, the treaty had been ratified by 105 states. The four
most prominent non-parties are the United States, Russia, India and China.
After long negotiations, the state delegates present in Rome decided to include
four categories of crimes under the remit of the ICC (Art. 5:1): genocide, crimes
against humanity, war crimes and, at least in principle, aggression. These crimes are
considered so heinous that their punishment should be a matter of international
5
concern and jurisdiction. The list of crimes can be added to at a review conference, to
be held seven years after the coming into force of the Rome Statute, i.e. in 2009.
Currently, it seems that state parties have little appetite for amending the list of
crimes. In fact, it is even doubtful whether they will give full effect to the aggression
category. This crime was formally included in the Statute, but essentially to no effect
since the negotiators at the Rome Conference could not agree on a definition of what
constitutes aggression and how it is to be prosecuted, leaving these essential details to
future amendments instead (Art. 5:2).
According to Art. 6, genocide covers ‘acts committed with intent to destroy, in
whole or in part, a national, ethnical, racial or religious group’. Of note, much like the
1948 Convention on the Prevention and Punishment of the Crime of Genocide, the
Rome Statute does not cover political and social groups within the scope of genocide,
which according to Schabas (2004: 40) has been frequently criticized. Art. 7 of the
Statute lists 11 acts of crimes against humanity, from the somewhat anachronistic
crime of apartheid to the widely recognized crimes of murder, torture and enforced
disappearance as well as ‘rape, sexual slavery, enforced prostitution, forced
pregnancy, enforced sterilization’, only very recently recognized as crimes against
humanity. Prosecution of such acts is always subject to the requirement that they are
‘committed as part of a widespread or systematic attack directed against any civilian
population, with knowledge of the attack’ (Art. 7:1). Art. 8 of the Statute represents
perhaps one of the greatest steps forward. Very explicitly, it extends the scope of war
crimes from international conflicts to civil wars (Schabas 2004: 54). It covers ‘grave
breaches of the Geneva Conventions’ together with ‘other serious violations of the
laws and customs applicable in international armed conflict’, but also ‘serious
violations of article 3 common to the four Geneva Conventions’ as well as ‘other
6
serious violations of the laws and customs’ for civil wars. However, state parties can
opt out of the provisions for war crimes for a one-off transitory period of seven years
(Art. 124).
The Court can exercise jurisdiction over individuals whether or not their nation-
state is party to the ICC as long as the crime has been committed within the territory
of state parties. However, the Court requires the consent of either the state of the
accused national or the state, in which the crime took place, unless the case is referred
to the Court by the UN Security Council. Also, the Court is only meant to take action
under the principle of complementarity, i.e. only when a state ‘is unwilling or unable
genuinely to carry out the investigation or prosecution’ (Art. 17). Commanders can be
prosecuted for crimes committed by forces under his or her command if the
commander knew about the crimes and could have prevented them (Art. 28).
Conversely, offenders cannot take recourse to the argument that they committed
crimes upon superior orders (Art. 33). No offence can be prosecuted that took place
before the Rome State came into effect in July 2002 (Art. 11).
3. A potential moral hazard problem: the ICC as an excuse for inaction?
Most extant scholarship has focused on trying to make sense of US resistance toward
the ICC and has discussed at length the credibility and merits of the two main US
objections toward the Court – see, for example, Forsythe (2002), Mayerfeld (2003),
Fehl (2004), Schabas (2004), Weller (2004), Ralph (2005) and Johansen (2006). The
two main explicitly stated objections are the potential for abuse of the Court by enemy
states and the risk to US military personnel on active duty abroad (Bolton 2002).
Clearly, the Rome Statute could have been constructed in a way that would have
made it easy for the US to ratify. In particular, it could have allowed permanent
7
members of the UN Security Council to veto any investigation and prosecution, thus
effectively enabling them to protect any of their nationals or those of their allies.
Doing so would have meant creating large “loopholes” in exchange for the benefit of
support by the US.
This proved unacceptable to the vast majority of negotiating states, however.
Instead, the parties chose an option that risked US withdrawal from the entire project.
The UN Security Council can defer investigations and prosecutions repeatedly for a
period of twelve months each (Art. 16), but none of its permanent members has the
power to veto the initiation of such proceedings in the first place. As a result, the Bush
administration withdrew Clinton’s signature under the Rome Statute and it began to
actively undermine the ICC by, first, refusing to support UN peacekeeping missions
unless US citizens are explicitly exempted from ICC proceedings and, second,
pressurizing state parties via the threat of economic and political sanctions into
concluding immunity agreements guaranteeing that no US citizens will be held or
extradited for the purpose of ICC proceedings (Johansen 2006). However, the Bush
administration has recently wavered a bit in its resolution to refuse co-operation with
the ICC. In particular, it has abstained from vetoing a UN Security Council vote
referring the Darfurian crisis in Sudan to investigation by the ICC in March 2005
(Grono 2006: 629).
The decision by the state parties to solve the crucial issue of the role and power
of permanent UN Security Council members in the particular way described above
has prompted critics to question the enforcement capacity of the ICC. Fehl (2004:
358), for example, wonders ‘…why was the new institution designed in such a way
that it failed to gain the support of the one country which is, by all accounts, most
important for enforcing its future decisions?’. This is a relevant question to ask, after
8
all the US is in a unique position due to its quasi-universal deployment of troops and
infrastructure and its logistical and military capacity to intervene virtually anywhere
in the world.
Yet, this critique only scratches the surface of a potential problem that reaches
much further than the enforcement of future Court decisions. The very creation of the
ICC might present a fundamental problem, namely via the negative effect it
potentially has on humanitarian intervention and multinational peacekeeping. There
are two reasons why the ICC might have such a negative effect. One argues with a
direct deterrence effect, the other one with a potential moral hazard problem. The
former is a favourite argument of the US administration used in its opposition to the
Court. Let us briefly discuss each argument in turn.
The ICC may directly deter humanitarian intervention and peacekeeping if
those who intervene and contribute to peacekeeping must fear that their military
personnel may face future indictment for their actions by the ICC. As David Scheffer,
then US Ambassador-at-Large for War Crimes Issues has put it in addressing the UN
General Assembly in 1998: ‘The Rome treaty will become the single most effective
brake on international and regional peacekeeping in the 21st century.’ (cited in
Schmitt and Richards 2000). Similarly, US Major Michael L. Smidt (2001: 240)
contends that ‘the court will likely make the world a much more dangerous place
because it will likely deter the forces of good, which will allow the forces of evil to
act with impunity’. India, another prominent critic of the ICC and, contrary to the US,
a troop contributor to many multinational peacekeeping missions, has similarly raised
the issue of a negative effect of the ICC on peacekeeping operations and concern for
its army personnel involved in such operations as part of its case against ICC
ratification (Koshy 2004: 2439; McDoldrick 2004: 440).
9
Such arguments are rather implausible, however. It is possible, of course, that
members of an intervening or peacekeeping force will commit crimes that fall under
the auspices of the ICC and might be indicted for them. But judged from past
experience, committing such crimes during interventions or peacekeeping missions is
the rare exception, not the rule, and would constitute cases that should be prosecuted
by the state to which these soldiers belong to itself, such that the ICC would never
become active in the first place. Neither is it likely that the ICC would become
politicised to enact dubious prosecutions against intervening forces. That the
International Criminal Tribunal for the former Yugoslavia (ICTY) quickly dismissed
charges against North Atlantic Treaty Organization (NATO) countries provides a case
in point. Rather, what the US seems to be concerned about is that the Court might
become active in the case of military interventions for which the humanitarian
justification is very dubious, like the invasions of Grenada and Panama in 1983 and
1989, respectively. Even then, however, it is difficult to see how the Court would
become active in the absence of a detailed provision for crimes of aggression.
The moral hazard problem is more pertinent and a priori more plausible. As
Smith (2002: 177) has formulated it succinctly: ‘If international actors feel confident
that human rights criminals will eventually be brought to justice, either in their own
countries or before the ICC, they may be less inclined to intervene to stop human
rights crimes while they are happening, something international actors have been
reluctant to do in any case.’ As a consequence, ‘the ICC may become a virtuous
excuse for states to turn a blind eye to atrocities, a moral free ride on the coattails of
humanitarian law’ (Smith 2002: 178).
Such suspicion and criticism is fuelled by the reaction of the international
community to the crimes committed during the Yugoslav wars and the Rwandan
10
genocide. In both cases, states arguably failed to intervene or intervened far too late.
In both cases, states tried to make up for their failure by installing special ad hoc
tribunals to prosecute crimes ex post. Aryeh Neier (1998: 112), co-founder and then
director of Human Rights Watch, argues that the Yugoslav tribunal acted as a
substitute for action: ‘Facing domestic criticism for allowing the slaughter to continue
unchecked, some governments seemed to feel obliged to show that they were doing
something’, and without much actual political cost. Similarly, Forsythe (1994: 403)
has commented that European states ‘felt the need to give the appearance of doing
something about violations of humanitarian law in the former Yugoslavia’, but
‘lacking the political will to act decisively to curtail abuses of prisoners and civilians,
they endorsed or went along with the creation of the Tribunal’. Is something similar
happening with the ICC?
4. An Indirect Test for the Moral Hazard Problem
The potential moral hazard problem created by the ICC cannot be directly observed or
tested for. As an indirect test, we estimate the determinants of its ratification. The idea
is that if the ICC creates an excuse for states which are reluctant to act, then these
states should be the prime supporters of the Court, which should reveal itself via
(early) ratification. In other words, if the ICC represented a real moral hazard
problem, then states which have been relatively inactive in the past as interveners in
foreign humanitarian crises should be the prime supporters of the ICC. If these states
were the prime supporters of the ICC then this would suggest that the court functions
as a substitute for action (the moral hazard problem is very relevant). If the opposite
were the case – that is, if states which have been more active in the past were its
11
prime supporters – then this would suggest that the ICC functions as a complement to
action, which would imply that the moral hazard problem is rather irrelevant.
For this argument to be valid, it must not be the case that countries which have
been active in the past ratify (early on) precisely because they are anxious to get the
ICC established such that they have an excuse for being less active in the future. I
contend that this is a safe presumption to make for either one of two reasons. First, if
preferences for intervention in foreign humanitarian crises are allowed to differ across
countries, then those countries which have been more active in the past may have a
stronger taste for intervention than less active countries and since preferences have
not changed will keep having this stronger taste. Second, if preferences for
intervention are assumed to be equal across countries, then the side constraints they
face will still differ. If domestic public pressure has managed to induce certain
countries to be more active in the past than others even in the absence of an ICC, then
it is likely that this same domestic public will manage to impose enough pressure on
its government to remain active in the future as well even after the ICC has been
established since the domestic pressure groups certainly regard the ICC as a
complement to action, not as a substitute.1 It is the countries, which have managed to
withstand pressure for action in the past that are likely to be tempted by the potential
moral hazard created by the ICC. For these countries, the establishment of an ICC
could potentially represent a means to deflect criticism toward its inaction, thus
enabling them to remain inactive in the future.
This is not to say that the establishment of the ICC could not represent a moral
hazard in a specific case even for a country, which has been a rather active intervener
1 See, for example, the NGO “Coalition for an International Criminal Court” website at
http://www.iccnow.org/.
12
in the past. Even a generally active country may use the existence of the ICC as a
means to deflect criticism toward its inaction due to the particular contextual
circumstances of the specific case at hand (for example, because the country is on
friendly terms with the offending government). However, one would expect that other
more active countries would be willing to step in, unless intervention similarly goes
against their interests. What this means, in sum, is the following: If states which have
been relatively inactive in the past are the prime supporters of the ICC, then this
would suggest that the moral hazard is very relevant, whereas the moral hazard is less
relevant if it is mainly supported by countries that have been rather active in the past.
But even then, the moral hazard still exists and may, depending on the circumstances
of the specific case, represent more or less of a problem.
A final critical hurdle for the indirect test of the moral hazard problem to work
is the presumption that countries are generally under pressure to act in foreign
humanitarian crises. If there is no such pressure, then there is no need to deflect
criticism for inaction and the ICC cannot exert any moral hazard for countries
unwilling to act. The liberal democracies of developed countries of the world are
more likely to experience such pressure domestically than developing countries due to
their stronger civil society and due to their commitment to human rights protection
(Andersson 2002; Lebovic 2004). However, even developing countries can be under
pressure from the outside. Several observers have attributed the increasing role of
developing countries in post-Cold War peacekeeping missions to pressure from
developed countries (see, for example, Byman and Waxman 2002; Pugh 2004;
Neethling 2004). The main result of this article’s analysis upholds if the sample is
restricted to developed countries only. It is thus robust to excluding developing
13
countries from the sample, for which the moral hazard problem may be less of an
issue in the first place.
5. Research Design
As mentioned already, our indirect test of the moral hazard problem works via
estimating the determinants of ratification of the Rome Statute. To be precise, I
estimate the determinants of ratification delay with an event type model. Such a
model uses more information than a pure cross-sectional probit or logit model with a
simple ratification dummy as dependent variable. The extra information used is the
time delay between the opening of the Rome Statute for signature and ratification
until eventual ratification, for those countries that have ratified. The event type model
thus not only distinguishes between ratifiers and non-ratifiers, but it further
distinguishes among the ratifying countries between early and late ratifiers, assuming
that early ratification reveals stronger support for the ICC than late ratification. This
can be justified on two accounts: First, early ratification increases the probability that
an international treaty will come into force as typically a minimum threshold of
ratifications needs to be exceeded (60 in the case of the Rome Statute). Second, the
more countries ratify early on, the more credibility the treaty gains and the higher the
pressure on remaining states to ratify the treaty as well. Strong supporters of a treaty
should thus not merely ratify a treaty, but will be keen to ratify early on.
Data on the year of ratification, acceptance, approval or accession of the Rome
Statute is taken from United Nations (2007). The country year rather than country day
forms the basis of our sample since the control variables are only available on an
annual basis. Smith (2004) argues that there are distinct regional norms of state
participation in the ICC following from preference heterogeneity across regions. To
14
account for this possibility, we use a stratified Cox proportional hazard model, where
we stratify by regions following largely World Bank country classification.2 This
allows each region to have its own baseline hazard. Formally, the dependent variable
is the conditional probability ρ(t) that ratification occurs at time t given that the
country has not ratified before t; this is the hazard of ratification:
ρi(t) = ρ0i(t)exp(Tx(t)), (1)
where i stands for the ith stratum (region), ρ0i(t) is the “baseline hazard”,
differing between regions (but uniform within a region), and T is the vector of
parameters to be estimated. The time-variant underlying baseline hazard of
ratification may depend on unobserved variables, possibly in complex ways. An
advantage of the Cox model is that the baseline hazard does not need to be estimated.
The likelihood function is constructed using the observation that the probability that
country i ratifies at time ti equals
ijij ttjii
Tii
T
ttjij
iiii t
tt
tt
||
))(exp())(exp(
)()(
)(̂xβ
xβ
. (2)
The likelihood function to be maximized with respect to the vector T then
equals it ii t )(̂ .
The two central explanatory variables are military intervention abroad and
contributions to peacekeeping. I use the Uppsala/PRIO dataset as source for the
variable measuring intervention in foreign civil conflicts (Gleditsch et al. 2002). The
specific measure used is the sum of country years in which a country intervened in a
foreign civil conflict since 1990, weighted by conflict intensity. The decision to use
2 The difference is that Canada and the US form one region together with Central America and the
Caribbean and South American countries constitute their own region.
15
1990 as a cut-off point is to account for the fact that the world fundamentally changed
after the end of the Cold War. The measure is weighted by conflict-intensity since it
takes much more courage and commitment to intervene in more intense foreign
conflicts. No attempt was made to distinguish “humanitarian” intervention from “self-
interested” interventions since there is no objective way of doing so. Data on
contributions to peacekeeping are taken from a database of the Stockholm
International Peace Research Institute (SIPRI), available at
http://conflict.sipri.org/SIPRI_Internet/. The specific measure used is the sum of
country years in which a country contributed troops to a multinational peacekeeping
operation abroad since 1990, whether led by the UN, by a regional inter-governmental
organization or by an ad hoc group of states.
There are six control variables. Democracy is measured by Polity data and
accounts for the fact that democracies are much more likely to ratify international
human rights treaties (Landman 2005). Per capita income accounts for the possibility
that demand for the ICC may be a normal good, using data from World Bank (2006).
The remaining control variables are meant to capture the expected future potential
costs of ratification to the ratifying state, its leaders and citizens. Note that since no
offences prior to the entry into force of the Rome Statute can be prosecuted and the
Court will exert its major impact only in years to come, the variables we use
approximate the future costs of ratification by looking at potentially offensive
behaviour in the past. This will be a good proxy for expected future costs of
ratification only if one is willing to make the assumption that states which have
engaged in potential offences in the past (or rather their political leaders) have a taste
for and are therefore likely to commit such potential offences in the future.
16
To start with, we include a variable measuring the sum of country years since
1990, in which a genocide or politicide has taken place in a country, weighted by the
intensity of the crime, with data taken from Harff (2006).3 Despite the fact that the
Rome Statute covers only genocides explicitly, not politicides, we do not exclude
politicides from this measure since they are likely to fall under the category of crimes
against humanity. Empirically, it makes little difference to our estimations if we
exclude politicides.
To our knowledge, there is no data measuring war crimes. In its absence, we use
data on conflict involvement. Put simply, the argument is that nationals from a
country that is involved in more and more intense armed conflicts are more likely to
be indicted for war crimes. The link may be weak, but we see no other way of
capturing this aspect. The specific variables we use are the sum of country years since
1990, in which a country was involved in an international or civil armed conflict,
weighted by conflict intensity, using again data from the Uppsala/PRIO project. The
last control variable is the number of times a state has initiated an international armed
conflict since 1990, using and slightly extending data from Gleditsch (2004). Whilst,
as discussed above, aggression is not a fully specified crime under the auspices of the
Rome Statute yet, this may change in the future, which may deter states with a
preference for initiating international conflicts from ratifying.
3 Genocide and politicide are defined as events involving ‘the promotion, execution, and/or implied
consent of sustained policies by governing elites or their agents or in the case of civil war, either of the
contending authorities that result in the deaths of a substantial portion of a communal group or
politicized non-communal group. In genocides the victimized groups are defined primarily in terms of
their communal (ethnolinguistic, religious) characteristics. In politicides, by contrast, groups are
defined primarily in terms of their political opposition to the regime and dominant groups.’
17
6. Results
Table 1 presents the Cox regression results. To start with, the sample includes both
developed and developing countries. The military intervention and peacekeeping
variables have positive and statistically significant coefficients in columns I and II,
respectively, when entered separately to the control variables. They remain significant
when entered together in column III. Countries that intervene more in foreign civil
conflicts and contribute more to multinational peacekeeping abroad are therefore
statistically significantly more likely to have ratified the Rome Statute (early on).
< Insert Table 1 around here >
With respect to the control variables, democracy is a clearly statistically
significant positive determinant of ratification, but per capita income has no effect. A
history of initiating wars deters countries from ratifying, as does involvement in
international armed conflicts (the coefficient is marginally insignificant in column I).4
Experience of civil war does not reduce the likelihood of a country ratifying. Neither
does a history of genocide and politicide.
In columns IV to VI we repeat the set of estimations for a sample of only
developed countries (member countries of the Organisation of Economic Co-
operation and Development or the European Union).5 As mentioned before, this is to
account for the fact that the pressure to become active may be far lower in developing
4 Goodliffe et al. (2006) similarly find that the expected costs of ratification deter some countries from
ratifying.
5 The genocide/politicide history variable is dropped since no developed country experienced such an
event during the relevant time period.
18
countries compared to developed countries, such that the moral hazard may be less
relevant in developing countries in the first place. Despite the substantial loss of
observations and therefore of efficiency of estimation, the results show that both the
intervention in foreign civil conflicts and the contribution to multinational
peacekeeping variables remain statistically significant if entered on their own. If
entered together, the peacekeeping variable is still significant, whereas the
intervention variable is not. This is likely to be due to the high correlation between the
two variables (r = 0.62 in this sample). A chi-squared test rejects the hypothesis that
both variables are equal to zero at p > 0.0057.
7. Conclusion
The ICC represents a milestone for the international human rights movement. For the
first time in the history of humankind, there exists the general possibility of
prosecuting political leaders as well as ordinary citizens who have committed crimes
of genocide, war crimes as well as crimes against humanity before an international
court with quasi-universal jurisdiction. The ICC currently investigates allegations in
four African countries – Central African Republic, Democratic Republic of Congo,
Sudan and Uganda – and it has initiated its first trial against Congolese national
Thomas Lubanga Dyilo, indicted for war crimes.
Critics have argued, however, that the establishment of the ICC may create a
new moral hazard. States reluctant to intervene militarily and to contribute to
multinational peacekeeping may use the existence of the ICC as an excuse for their
inaction. It may allow states to reassure their domestic publics that something is done
and that crimes can no longer be committed with impunity, without incurring the
substantial costs for preventing those crimes in the first place. The vague promise of
19
ex post justice would then be likely to replace immediate protection. The implication
of this would be that the ICC on the whole could represent a step back rather than
forward in the quest for protecting victims of genocide, war crimes and other crimes
against humanity.
The existence and actual relevance of this possibility for a new moral hazard
problem is difficult to ascertain directly. This article has offered an indirect test.
Specifically, if the moral hazard argument applies, then those states which have been
rather reluctant to intervene militarily and to contribute to multinational peacekeeping
in the past should also be the prime supporters of the ICC and therefore be the first
ones to ratify its Rome Statute. However, we find the exact opposite. Those states that
are willing to play an active role in interventions and peacekeeping are also, on
average, the prime supporters of the ICC. It would therefore appear that the ICC is not
a sorry excuse for inactive states, but another strategy for active states to combat
situations in which acts of genocide, war crimes as well as crimes against humanity
can happen and can be committed with impunity.
Yet, given the indirectness of our test for the moral hazard created by the ICC,
only the future will tell whether the ICC will complement or substitute action. The
current situation in Darfur, Sudan, does not provide a clear picture. On the one hand,
several hundred thousand people have already reportedly been killed and while there
is the will from several, mainly Western and African states to intervene in a
meaningful way, this has yet to happen. The case is under ICC investigation following
a referral by the UN Security Council, but it appears that this move was part of a
strategy to increase pressure on the Sudanese government to facilitate real protection
of people at risk rather than a substitute for intervention.
20
Another caveat is that there remains the possibility that the existence of the ICC
creates an incentive even for a state, which has been rather active in the past, not to
intervene in a specific case if intervention goes against the state’s interests. That is,
even if in general the ICC does not represent a moral hazard, it can do so for specific
countries in specific cases. However, on the whole the fact that the ICC is primarily
supported by states, which have been active in the past, suggests that it will function
as a complement to, not a substitute for action. If accompanied by action, the ICC can
form an important step forward in the protection of victims. Our estimation results
paint a cautiously optimistic picture in demonstrating that the ICC has been ratified
early on by states, which have been willing to act in the past. This bodes well for the
ICC and its future contributing role to the fight against heinous political crimes,
despite continued opposition to the Court from the US in particular, but many other
countries as well, including China and India.
21
References
Andersson, A. 2002. ‘United Nations intervention by united democracies? State
commitment to UN interventions 1991-99’, Cooperation and Conflict 37: 363-
386.
Annan, Kofi. 1997/98. ‘Advocating for an International Criminal Court’, Fordham
International Law Journal 21: 363-366.
Bolton, John R. 2002. The United States and the International Criminal Court.
Washington DC: U.S. State Department. http://www.state.gov/t/us/rm/15158.htm
Byman, D. and M. Waxman, M. 2002. The dynamics of coercion: American foreign
policy and the limits of military might. Cambridge: Cambridge University Press.
Fehl, Caroline. 2004. ‘Explaining the International Criminal Court: A “Practice Test”
for Rationalist and Constructivist Approaches’, European Journal of
International Relations 10 (3): 357-394.
Forsythe, David P. 2002. ‘The United States and International Criminal Justice’,
Human Rights Quarterly 24: 974-991.
Gleditsch, Kristian, 2004. ‘A Revised List of Wars Between and Within Independent
States, 1816-2002’, International Interactions 30: 231-262.
Gleditsch, Nils Petter; Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg, and
Håvard Strand. 2002. ‘Armed Conflict 1946–2001: A New Dataset’, Journal of
Peace Research 39 (5):615-37.
Goodliffe, Jay, Darren G. Hawkins, Christine Horne, and Daniel L. Nielson. 2006.
Norm Enforcement, Dependence Networks and the International Criminal Court.
Mimeo. Brigham Young University and Washington State University.
Grono, Nick. 2006. ‘Briefing – Darfur: The International Community’s Failure to
Protect’, African Affairs 105 (421): 621-631.
22
Harff, Barbara. 2006. Genocide and Politicide Project.
http://www.cidcm.umd.edu/inscr/genocide/
Johansen, Robert C. 2006. ‘The Impact of US Policy toward the International
Criminal Court on the Prevention of Genocide, War Crimes, and Crimes Against
Humanity’, Human Rights Quarterly 28: 301-331.
Koshy, Ninan. 2004. ‘International Criminal Court and India’, Economic and Political
Weekly 39 (23): 2439-2440.
Landman, Todd. 2005. Protecting Human Rights: A Comparative Study. Washington
DC: Georgetown University Press.
Lebovic, J. H. 2004. ‘Uniting for peace? Democracies and United Nations peace
operations after the Cold War’, Journal of Conflict Resolution 48: 910-936.
Mayerfel, Jamie. 2003. ‘Who Shall be Judge?: The United States, the International
Criminal Court, and the Global Enforcement of Human Rights’, Human Rights
Quarterly 25: 93-129.
McGoldrick, D. 2004. ‘National Implementation and Political Responses’, in Dominic
McGodlrick, Peter Rowe and Eric Donnelly, eds., The Permanent International
Criminal Court – Legal and Policy Issues. Oxford: Hart Publising (337-452)
Neethling, T. 2004. ‘International peacekeeping trends: The significance of African
contributions to African peacekeeping requirements’, Politikon 31: 49-66.
Neier, Aryeh, 1998. War Crimes. New York: Random House.
Pugh, M. 2004. ‘Peacekeeping and critical theory’, International Peacekeeping 11:
39-58.
Ralph, Jason. 2005. ‘International Society, the International Criminal Court and
American Foreign Policy’, Review of International Studies 31: 27-44.
23
Schabas, William A. 2004. ‘United States Hostility to the International Criminal
Court: It’s all About the Security Council’, European Journal of International
Law 15 (4): 701-720.
Schabas, William A. 2004. An Introduction to the International Criminal Court.
Second Edition. Cambridge: Cambridge University Press.
Schmitt, Michael N., & Peter J. Richards, 2000. ‘Into Unchartered Waters – The
International Criminal Court’, Naval War College Review LIII (1).
Smidt, Michael L. 2001. ‘The International Criminal Court: an Effective Means of
Deterrence?’, Military Law Review 167: 156-240.
Smith, Heather M. 2004. The International Criminal Court and Regional Diffusion.
Paper presented at the Fifth Pan European Conference on International Relations,
The Hague. http://www.humanrights.uconn.edu/rese_papers/HeatherSmith.pdf
Smith, Thomas W. 2002. ‘Moral Hazard and Humanitarian Law: The International
Criminal Court and the Limits of Legalism’, International Politics 39: 175-192.
United Nations. 2007. Ratification Status of the Rome Statute of the International
Criminal Court.
http://untreaty.un.org/ENGLISH/bible/englishinternetbible/partI/chapterXVIII/tr
eaty11.asp
Weller, Marc. 2002. ‘Undoing the Global Constitution: UN Security Council Action
on the International Criminal Court’, International Affairs 78 (4): 693-712.
World Bank. 2006. World Development Indicators on CD-Rom. Washington, DC:
World Bank.
24
Table 1. Ratification of the Rome Statute to the International Criminal Court.
I(all)
II(all)
III(all)
IV(developed)
V(developed)
VI(developed)
Democracy 0.119 0.103 0.104 0.045 0.032 0.080(5.18)*** (4.51)*** (4.50)*** (0.28) (0.20) (0.48)
GDP per capita 0.000 -0.000 -0.000 -0.000 -0.000 -0.000(0.14) (0.22) (0.31) (0.71) (1.41) (1.30)
Genocide/politicide history -0.118 -0.079 -0.090(0.51) (0.53) (0.51)
International war history -0.129 -0.152 -0.178 -0.295 -0.046 -0.411(1.56) (1.80)* (2.04)** (0.64) (0.11) (0.88)
Civil war history -0.026 -0.037 -0.030 -0.169 -0.222 -0.205(0.89) (1.22) (0.98) (2.16)** (2.85)*** (2.43)**
War initiation history -0.765 -0.689 -0.941 -0.423 -0.564 -0.441(1.88)* (1.77)* (2.41)** (0.72) (1.01) (0.78)
Intervention in foreign civil wars history 0.157 0.118 0.239 0.152(2.38)** (1.78)* (1.97)** (1.15)
Multinational peacekeeping history 0.021 0.019 0.030 0.026(2.52)** (2.16)** (2.68)*** (2.09)**
# of countries 151 151 151 35 35 35of which have ratified in study period 75 75 75 30 30 30
Observations 609 609 609 117 117 117Log pseudolikelihood -201.9 -200.8 -200.1 -54.5 -53.7 -53.4
Analysis is by Cox proportional hazard estimation, stratified by regions. Absolute z-values in parentheses. ***, **, and * indicate significance at
the .01, .05, and .10 levels, respectively.